Role of VTA dopamine neurons and neuroligin 3 in sociability traits related to nonfamiliar conspecific interaction

(1)

Article

Reference

Role of VTA dopamine neurons and neuroligin 3 in sociability traits related to nonfamiliar conspecific interaction

BARISELLI, Sebastiano, et al.

Abstract

Atypical habituation and aberrant exploration of novel stimuli have been related to the severity of autism spectrum disorders (ASDs), but the underlying neuronal circuits are unknown. Here we show that chemogenetic inhibition of dopamine (DA) neurons of the ventral tegmental area (VTA) attenuates exploration toward nonfamiliar conspecifics and interferes with the reinforcing properties of nonfamiliar conspecific interaction in mice. Exploration of nonfamiliar stimuli is associated with the insertion of GluA2-lacking AMPA receptors at excitatory synapses on VTA DA neurons. These synaptic adaptations persist upon repeated exposure to social stimuli and sustain conspecific interaction. Global or VTA DA neuron-specific loss of the ASD-associated synaptic adhesion molecule neuroligin 3 alters the behavioral response toward nonfamiliar conspecifics and the reinforcing properties of conspecific interaction.

These behavioral deficits are accompanied by an aberrant expression of AMPA receptors and an occlusion of synaptic plasticity. Altogether, these findings link impaired exploration of nonfamiliar conspecifics to VTA DA neuron [...]

BARISELLI, Sebastiano, et al . Role of VTA dopamine neurons and neuroligin 3 in sociability traits related to nonfamiliar conspecific interaction. Nature Communications , 2018, vol. 9, no.

1, p. 3173

DOI : 10.1038/s41467-018-05382-3 PMID : 30093665

Available at:

http://archive-ouverte.unige.ch/unige:111209

Disclaimer: layout of this document may differ from the published version.

(2)

Role of VTA dopamine neurons and neuroligin 3 in sociability traits related to nonfamiliar conspeci ﬁ c interaction

Sebastiano Bariselli

¹

, Hanna Hörnberg

²

, Clément Prévost-Solié

¹

, Stefano Musardo

¹

, Laetitia Hatstatt-Burklé

²

, Peter Scheiffele

²

& Camilla Bellone

¹

Atypical habituation and aberrant exploration of novel stimuli have been related to the severity of autism spectrum disorders (ASDs), but the underlying neuronal circuits are unknown. Here we show that chemogenetic inhibition of dopamine (DA) neurons of the ventral tegmental area (VTA) attenuates exploration toward nonfamiliar conspeci ﬁ cs and interferes with the reinforcing properties of nonfamiliar conspeci ﬁ c interaction in mice.

Exploration of nonfamiliar stimuli is associated with the insertion of GluA2-lacking AMPA receptors at excitatory synapses on VTA DA neurons. These synaptic adaptations persist upon repeated exposure to social stimuli and sustain conspeci fi c interaction. Global or VTA DA neuron-speci fi c loss of the ASD-associated synaptic adhesion molecule neuroligin 3 alters the behavioral response toward nonfamiliar conspeci fi cs and the reinforcing properties of conspeci fi c interaction. These behavioral de fi cits are accompanied by an aberrant expression of AMPA receptors and an occlusion of synaptic plasticity. Altogether, these findings link impaired exploration of nonfamiliar conspecifics to VTA DA neuron dysfunction in mice.

DOI: 10.1038/s41467-018-05382-3

OPEN

1Department of Basic Neurosciences, University of Geneva, 1211 Geneva, Switzerland.²Biozentrum of the University of Basel, 4056 Basel, Switzerland. These authors contributed equally: Sebastiano Bariselli, Hanna Hörnberg, Clément Prévost-Solié. Correspondence and requests for materials should be addressed to C.B. (email:Camilla.Bellone@unige.ch)

1234567890():,;

(3)

F rom infancy, we encounter an array of diverse stimuli from the environment. Repeated exposure to a stimulus can result in habituation whereas nonfamiliar stimuli usually increases exploratory behavior. Habituation and novelty recognition allow us focusing attention on what is unknown, promote exploratory behavior, facilitate learning, and are predictive of cognitive function later in life

¹

. Several neuropsychiatric disorders are characterized by deﬁcits in habituation and novelty exploration.

In autism spectrum disorder (ASD), young patients show pro- longed attention to depictions of objects, but reduced attention to social stimuli

²

. Moreover, ASD patients are hyporesponsive to novel visual stimuli and exhibit slowed habituation to faces

^3,4

. Such alterations are observed in a signiﬁcant number of indivi- duals with ASD, as they have been reported in clinical studies using diverse stimuli and read-outs

^5–7

. However, the circuits and neuronal mechanisms underlying this speciﬁc aspect of the ASD phenotype remain largely unknown.

Dopamine (DA) neurons in the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc) may contribute to the habituation to familiar stimuli and to the exploration of non- familiar stimuli. DA neurons increase their activity in response to novel environments

⁸

, to stimuli of positive or negative value

⁹

, and to natural rewards

¹⁰

. Interestingly, these neurons also respond to nonrewarding novel stimuli and their responses habituate when the stimulus becomes familiar

^11,12

. This has led to the proposal that novelty by itself may be rewarding. In rodents, nonfamiliar conspeciﬁcs or nonfamiliar objects increase Ca

²⁺

-transients in VTA DA neurons and this activity is necessary to promote social, but not object exploration

¹³

. Glutamatergic synapses onto DA neurons undergo several forms of synaptic plasticity that may contribute to the mod- iﬁcation of social interactions in response to experience. Spe- ciﬁc synaptic adaptations have been described during development, after drug exposure, cue-reward learning, reci- procal social interactions, and after repeated burst stimulation of DA neurons

¹⁴^–¹⁸

. Furthermore, glutamatergic transmission is altered in several ASD animal models

¹⁹

, and we have recently shown that deﬁcits in the postnatal development of excitatory transmission onto VTA DA neurons lead to sociability deﬁ- cits

²⁰

. Notably, several studies highlight decreased social reward processing in patients with ASD

^21,22

, and these alterations have been hypothesized to precipitate further developmental con- sequences in social cognition and communication

²³

. Whether speciﬁc forms of synaptic plasticity in the VTA are induced by exposure to nonfamiliar stimuli (novelty-induced synaptic plasticity), and whether aberrant plasticity associated with exploration of nonfamiliar conspeciﬁc in the VTA is related to the maladaptive responses in ASD mouse models is still largely unknown.

In this study, we parse the response to and the preference for nonfamiliar conspecifics as specific aspects of sociability con- trolled by DA neurons. We demonstrate that intact VTA DA neuron excitability is necessary to express a preference for non- familiar conspecifics but not for nonfamiliar objects. Additionally, we adopt a conditioned place preference protocol, based on interaction with familiar or nonfamiliar conspecific, to demon- strate that VTA DA neuron function underlies the reinforcing properties of social interaction. Mice lacking the ASD-associated synaptic adhesion molecule neuroligin 3 (Nlgn3) exhibit aberrant exploration of nonfamiliar conspecifics as well as deficit in habituation processing. These phenotypes are recapitulated by VTA DA neuron-specific down-regulation of Nlgn3. Finally, we discovered a form of novelty-induced synaptic plasticity at glu- tamatergic inputs onto VTA DA neurons that sustains conspecific interactions and is impaired in Nlgn3 KO and Nlgn3 VTA DA knockdown mice.

Results

VTA DA neurons and exploration of nonfamiliar conspeciﬁcs.

Mice have been reported to interact with their conspeciﬁcs, to habituate upon repeated contact with the same subject, and to exhibit increased exploration when subsequently brought into contact with a nonfamiliar mouse

²⁴

. To examine whether VTA DA neurons regulate exploration of nonfamiliar conspeciﬁcs, we examined the behavior of mice in which the inhibitory DREADD (hM4Di)

²⁵

or mCherry were virally expressed in DA neurons of the VTA (VTA::DA

^hM4Di

: AAV5-hSyn-DIO-hM4Di-mCherry or VTA::DA

^mCherry

: AAV5-hSyn-DIO-mCherry injected into DAT- Cre mice, Fig. 1a). Virus infusions led to mCherry expression in 50% of TH

⁺

(tyrosine hydroxylase, an enzyme necessary for DA synthesis) VTA neurons and in few (2%) of TH

⁺

cells in the neighboring substantia nigra pars compacta (SNc; Supplementary Fig. 1a), conﬁrming preferential targeting of the VTA. Applica- tion of the hM4Di ligand clozapine-n-oxide (CNO) decreased the neuronal excitability of VTA::DA

^hM4Di

neurons compared to VTA::DA

^mCherry

ex vivo (Supplementary Fig. 1b) and decreases DA release in striatal regions in vivo

²⁶

.

We then assessed the time spent in social interaction upon repeated exposure to the same mouse (habituation) and the subsequent response to a nonfamiliar conspeciﬁc (Fig. 1b). To compare between social and nonsocial stimuli, we also examined the behavioral responses to familiar and nonfamiliar objects (Supplementary Fig. 1c–f). When repeatedly exposed to the same mouse (Fig. 1c) or object stimulus (Supplementary Fig. 1c; s1 and o1, respectively), VTA::DA

^hM4Di

animals injected with vehicle show progressive reduction in the interaction with the stimuli over days. This process has been referred to as habituation

²⁷

. After four habituation days, the animals increased their exploratory behavior toward either a nonfamiliar social (s2;

Fig. 1d) or a nonfamiliar object stimulus (o2, Supplementary Fig. 1d) at day 5.

To study the role of VTA DA neurons in this behavioral trait, DA neuron excitability was decreased by intra-peritoneal (i.p.) injection of CNO in VTA::DA

^hM4Di

before the exposure to the nonfamiliar stimulus at day 5. VTA::DA

^hM4Di

animals decreased their exploratory behavior toward the nonfamiliar conspeciﬁc whereas control VTA::DA

^mCherry

mice treated with CNO showed unaltered stimulus exploration (s2, Fig. 1e, f). Interestingly, when exposed to a nonfamiliar object (o2), both VTA::DA

^hM4Di

and VTA::DA

^mCherry

animals treated with CNO exhibited an increased exploration (Supplementary Fig. 1e, f). Thus, reducing VTA DA neuron excitability speciﬁcally alters the exploration of a nonfamiliar conspeciﬁc, but not of a nonfamiliar object, suggesting a differential requirement of DA neuron activity for driving exploration of social and inanimate stimuli.

VTA DA neurons and preference for nonfamiliar conspeci ﬁ cs.

To assess the role of VTA DA neuron excitability in mediating the exploration of a nonfamiliar conspeciﬁc over an inanimate object or a familiar conspeciﬁc stimulus, VTA::DA

^hM4Di

and VTA::DA

^mCherry

mice were subject to the three-chamber test

²⁸

under vehicle and CNO conditions. The test was performed twice:

ﬁrst, animals received either vehicle or CNO and, after 1 week of washout, the test was repeated and the pharmacological treatment was counterbalanced (Fig. 2a). To monitor potential off target effects of CNO, we also included VTA::DA

^mCherry

mice treated with CNO as controls. During the task, test mice were given a choice between an object (o1) versus a nonfamiliar mouse (s1 or s3) and subsequently a choice between a familiar (second expo- sure to s1 or s3) versus a nonfamiliar conspeciﬁc (s2 or s4).

Previous studies deﬁne sociability in this assay as longer time

spent in the chamber with the same-sex nonfamiliar mouse rather

(4)

than in the chamber with the object, and more time spent snifﬁng the same-sex mouse rather than snifﬁng the object

^29,30

. Accord- ing to these criteria, VTA::DA

^hM4Di

mice treated with vehicle (Supplementary Fig. 2a, d), VTA::DA

^mCherry

mice treated with CNO (Supplementary Fig. 2b, e) as well as VTA::DA

^hM4Di

mice treated with CNO (Supplementary Fig. 2c, f) exhibited sociability.

We observed a decreased distance moved upon CNO-mediated reduction of DA neuron excitability (Supplementary Fig. 2g).

However, despite the reduced locomotion, mice still expressed social preference.

In the second phase of the task, we assessed preference for social novelty that was defined as follows: longer time spent in the chamber with the same-sex nonfamiliar rather than in the chamber with the familiar mouse and more time spent sniffing the same-sex nonfamiliar conspecific rather than sniffing the familiar mouse, particularly during the first 5 min of the test

²⁸

. While preference for social novelty was exhibited by VTA::

DA

^hM4Di

mice treated with vehicle (Fig. 2b, e) and by VTA::

DA

^mCherry

mice treated with CNO (Fig. 2c, f), it was absent in VTA::DA

^hM4Di

mice treated with CNO (Fig. 2d, g). CNO treated VTA::DA

^hM4Di

mice displayed a reduction in distance moved (Fig. 2h). Additionally, to compare preference for social novelty

across groups, we calculated a “social novelty index”, as time spent snifﬁng the nonfamiliar stimulus minus time spent exploring the familiar target, in the ﬁrst and last 5 min of the assay. We found that the social novelty index was reduced by CNO injections in VTA::DA

^hM4Di

mice compared to both CNO treated VTA::DA

^mCherry

and vehicle treated VTA::DA

^hM4Di

(Fig. 2i). Altogether, these ﬁndings indicate that reducing the excitability of DA neurons decreases the exploration of novel social stimuli, when given a choice between nonfamiliar and familiar conspeciﬁcs.

VTA DA neurons and nonfamiliar conspeciﬁc reinforcement.

To investigate whether nonfamiliar conspecific interactions are reinforcing in mice, we performed a conditioned place preference (CPP) task (modified from 31, 32). Briefly, test mice are housed with familiar mice throughout the protocol. After the Pre-TEST, we performed 4 days of repeated conditioning where wild-type (WT) mice learn to associate one compartment of the apparatus with the presence of either a familiar conspecific (familiar, f1), a nonfamiliar conspecific (s1) or a nonfamiliar object (o1) stimulus, while the other compartment is left empty (Fig. 3a, b). At day 5 (Post-TEST) the preference to explore the two compartments, in

6 weeks postinfection

VTA

SNc

TH DIO-hM4Di-mCherry DAPI

a b

1

s1 o1 or

2

s1 o1 or

3

s1 o1 or

4

s1 o1 or

5

s2 o2 or Habituation Nonfamiliar DAT-Cre

(4–7 weeks)

AAV5-DIO-hM4Di or

AAV5-DIO-mCherry 4–6 wks

15′ 15′ 15′ 15′ 15′

Days

Vehicle i. p.Vehicle i. p.Vehicle i. p.Vehicle i. p.

Vehicle / CNO i. p.

0 100 200 300 400

P = 0.0237 P = 0.0009 P < 0.0001 s1 s1 s1 s1

c

_VTA::DA^hM4Di_{- vehicle} N= 37

d

0 50 100 150 200

250 ^s1 ^s2

P = 0.0008

VTA::DA^hM4Di - vehicle N= 19

–10 0 10 20 30

Social novelty index (s2–s1#4)

P = 0.0494

CNO VTA::DA

hM4Di

VTA::DA

mCherr y

VTA::DA

hM4Di

Veh

P > 0.9999

18 19

P = 0.0059

P = 0.0069

f

N= 19 s1 s1 s1 s1 s2

0 50 100 150

P < 0.0001

P < 0.0001

VTA::DA^hM4Di (N= 18) VTA::DA^mCherry (N= 19)

e

Time (day)

Time (day) Time (day)

1 2 3 4 5

4 5

1 2 3 4

CNO

Interaction time (s)

Interaction time (s) Interaction time (s)

s1 o1

Social stimuli Object stimuli s2

/

o2

/

Fig. 1VTA DA neuron excitability controls exploration of nonfamiliar conspeciﬁc.aRepresentative images (low and high magniﬁcation) of immuno- staining experiments against tyrosine hydroxylase (TH) enzyme (in green) performed on midbrain slices of DAT-Cre mice infected with AAV5-DIO- hM4Di-mCherry (red). Scale bar: 1 mm and 100μm.bExperimental time-course for the habituation/nonfamiliar exploration task.cTime course of time interaction for VTA::DA^hM4Dimice treated with vehicle during habituation phase. Friedman test (x²(4)=32.94,P< 0.0001) followed by Dunn’s test for planned multiple comparisons.dGraph reporting the time interaction at day 4 with s1 and at day 5 with s2 for VTA::DA^hM4Dimice treated with vehicle.

Wilcoxon test (W=156).eTime interaction over days during the habituation/nonfamiliar exploration task (s1 and s2 are nonfamiliar conspeciﬁc stimuli presented at day 1–4 and 5, respectively) for VTA::DA^hM4Diand VTA::DA^mCherrymice treated with CNO. Repeated measures (RM) two-way ANOVA (time main effect:F(4,140)=12.38,P< 0.0001; virus main effect:F(1,35)=3.13, P=0.0854; time × drug interaction:F(4,140)=9.32,P< 0.0001) followed by Bonferroni post hoc test.fSocial novelty index calculated from VTA::DA^hM4Ditreated with vehicle, VTA::DA^hM4Diand VTA::DA^mCherryboth treated with CNO. Kruskal–Wallis test (K(3)=10.26,P=0.0059) followed by Dunn’s multiple comparisons test.Nindicates number of mice. Error bars report s.e.m

(5)

absence of any stimulus, was quantified and compared to Pre- TEST. While no significant preference was developed for the familiar conspecifics (Fig. 3c and Supplementary Fig. 3a), mice exhibited preference for the compartment associated with the nonfamiliar conspecifics (Fig. 3d and Supplementary Fig. 3b), and

an avoidance for the novel object stimulus associated chamber (Fig. 3e and Supplementary Fig. 3c). Interestingly, across con- ditioning sessions, we observed habituation to all the stimuli (Fig. 3f–h). However, when the time of interaction with the sti- mulus during the ﬁrst and the last day of conditioning was

10 minutes

s2 s1

CNO

VTA::DA^mCherry

0 100 200 300

VTA::DA^mCherry- CNO (N= 15) 0–5 min 5–10 min

P = 0.0022 P = 0.0070 P = 0.0072

P = 0.0527

0 100 200 300

VTA::DA^hM4Di- CNO (N= 18) 0–5 min 5–10 min

P = 0.0191 P = 0.0182 P = 0.9943

P = 0.0061

P = 0.5666 P = 0.0115

DAT-Cre (4–7 weeks)

AAV5-DIO-hM4Di

a

3-chamber test 4–6 wks

o1 s1 s2

10 min 10 min

3-chamber test

o1 s3 s4

10 min 10 min

s1 s3

Vehicle or CNO i.p.

b c d

0 20 40 60 80 100 120

Time sniffing (s)

VTA::DA^mCherry- CNO

P < 0.0001 P = 0.0947 P = 0.0008

f

AAV5-DIO-mCherry

0 100 200 300

Time in chamber (s)

0–5 min 5–10 min

P < 0.0001 P = 0.0055 P < 0.0001

P = 0.0771

VTA::DA^hM4Di- vehicle (N= 18)

e

0 20 40 60 80 100 120

Time sniffing (s)

0–5 min 5–10 min P = 0.0092 P < 0.0001 P > 0.9999

VTA::DA^hM4Di- vehicle

0 1000 2000 3000

Distance moved (cm)

VTA::DA hM4Di

VTA::DA mCherr

y

CNO Vehicle

VTA::DA hM4Di

VTA::DA mCherr

y

CNO Vehicle

0–5 min 5–10 min

P < 0.0001 P < 0.0001

P < 0.0001 P = 0.0018

P < 0.0001 P = 0.0007

h

1 wk

Time in chamber (s) Time in chamber (s)

0 20 40 60 80 100 120

Time sniffing (s)

VTA::DA^hM4Di- CNO

P = 0.9757 P = 0.3079 P = 0.2356

g

–50 0 50 100

Social novelty index

P = 0.0217 P = 0.0212

0–5 min

i

VTA::DA hM4Di

VTA::DA mCherr

y

CNO Vehicle

VTA::DA hM4Di

VTA::DA mCherr

y

CNO Vehicle P = 0.0082

P = 0.9477 P = 0.3692 Center

Vehicle or CNO i.p.

Counter balanced o1

s1

Object Social

Familiar s1

/

s3

Nonfamiliar s2

/

s4

Familiar Nonfamiliar

Center

Familiar FamiliarCenter FamiliarCenter FamiliarCenter FamiliarCenter

Familiar Familiar

/

s3

(N= 18) (N= 15) (N= 18)

Nonfamiliar Nonfamiliar Nonfamiliar Nonfamiliar Nonfamiliar

Nonfamiliar Nonfamiliar

5–10 min

0–5 min 5–10 min Familiar Familiar

(6)

analyzed, we observed a longer interaction with the nonfamiliar conspeciﬁcs compared to the other stimuli at either time point (Fig. 3i). These data suggest that a nonfamiliar conspeciﬁc remains salient over days and promotes contextual associative learning.

To assess the role of VTA DA neuron excitability in mediating the reinforcing properties of nonfamiliar conspeciﬁc interactions, both control VTA::DA

^mCherry

and VTA::DA

^hM4Di

received injections of CNO before each conditioning session and were treated with vehicle before the Post-TEST (Fig. 3j). Control VTA::

DA

^mCherry

but not VTA::DA

^hM4Di

mice developed a preference for the compartment associated with the nonfamiliar conspecifics (Fig. 3k and Supplementary Fig. 3d, e). These observations suggest that the excitability of DA neurons mediates both the interaction with nonfamiliar conspecifics as well as the acquisi- tion of nonfamiliar conspecific-induced contextual associations.

Altered conspeci ﬁ c interactions in Nlgn3

^KO

mice. Patients with ASD exhibit slowed habituation to faces

⁴

and are less responsive to social reward

²²

. Thus, we tested whether a deletion of Nlgn3 in mice, a category 2 (strong candidate) classiﬁed ASD-linked gene (http://gene.sfari.org)

³³^–³⁵

encoding a postsynaptic adhesion molecule

³⁶

, might result in deficits in exploration of nonfamiliar conspecifics and in the reinforcing properties of conspecific interaction. Global Nlgn3

^KO

mice

³⁷

exhibit reduced ultrasonic vocalization and social memory in male–female interactions as well as altered motor behaviors and olfaction

^38–41

. We examined the interaction time upon repeated exposure to a familiar mouse (habituation) and the subsequent response to a nonfamiliar conspeciﬁc (Fig. 4a). Nlgn3

^KO

mice exhibited overall lower interaction times, no signiﬁcant habituation, and lacked the increased response to nonfamiliar conspeciﬁcs seen in Wild Type (WT) littermates (Fig. 4b, c and Supplementary Fig. 4a–d).

However, Nlgn3

^KO

mice showed habituation, increased explora- tion of nonfamiliar objects (Fig. 4d, e) and preference for non- familiar objects in a novel object recognition task (Fig. 4f–h). This indicates that both novelty preference and memory for objects are unaltered. In addition to impaired response to nonfamiliar con- speciﬁcs, Nlgn3

^KO

mutants exhibit alterations in motor activity (Fig. 4i) and marble burying (Fig. 4j). In an olfactory dis- crimination test

⁴²

, Nlgn3

^KO

male mice showed normal response and habituation to a social odor (Supplementary Fig. 4e). How- ever, the mutant mice had a signiﬁcantly decreased response when subsequently presented to a second (novel) social odor (Supplementary Fig. 4e). To further examine conspeciﬁc inter- action in Nlgn3

^KO

mice, we tested the reinforcing properties of social interaction

^31,32

. When mice are conditioned in a

conditioned place preference paradigm with familiar mice, Nlgn3

^KO

mice did not develop a preference for the social com- partments, whereas WT mice did (Fig. 4k, l, and Supplementary Fig. 4f, g). These ﬁndings suggest that Nlgn3

^KO

mice exhibit altered social interactions and defects in social reward behaviors.

Nlgn3 loss-of-function in VTA DA neurons alters sociability.

The diverse alterations in social but also nonsocial behaviors in Nlgn3

^KO

mice, indicate that multiple different systems might contribute to their phenotype. To test whether any alterations are due to Nlgn3 functions in VTA DA neurons we generated microRNA-based knock-down vectors for conditional suppres- sion of Nlgn3 expression (Supplementary Fig. 5a, b). Cre- dependent AAV-based vectors were injected into the developing VTA of DAT-Cre mice at postnatal days 5–6 and mice were analyzed using a battery of behavioral tests (AAV2-DIO-miR

^Nlgn3

in DAT-Cre mice: VTA::DA

^NL3KD

, Fig. 5a, b, and see Supple- mentary Fig. 5c for off-target areas affected and Supplementary Fig. 5d, e for further controls). Notably, VTA::DA

^NL3KD

mice exhibited a similar impairment in reinforcing properties of con- speciﬁc interaction as the global Nlgn3

^KO

mice in the conditioned place preference paradigm (Fig. 5c, d, and Supplementary Fig. 5f, g) indicating that Nlgn3 downregulation in VTA DA neurons is sufﬁcient to mimic this aspect of the global Nlgn3

^KO

phenotype.

Furthermore, when repeatedly exposed to the same and subse- quently to a nonfamiliar conspeciﬁc, VTA::DA

^NL3KD

mice showed an overall reduction in social exploration and a blunted response to novel conspeciﬁc stimuli (Fig. 5e–g, and Supple- mentary Fig. 5h–k). At the same time, VTA::DA

^NL3KD

mice showed preference for novel objects in the novel object recogni- tion task (Fig. 5h–j). Thus, there is a speciﬁc requirement for Nlgn3 in VTA DA neurons for appropriate exploration of non- familiar conspeciﬁcs and for the reinforcing properties of social interaction. By contrast, motor activity, marble burying, and social olfaction that are altered in global Nlgn3

^KO

mice were not modiﬁed in the VTA::DA

^NL3KD

mutants (Fig. 5k, l, Supplemen- tary Fig. 5l). Interestingly, we observed that knock-down of Nlgn3 in VTA-DA neurons of adult mice produced a similar but less pronounced social interaction phenotype as in developing ani- mals, with reduced habituation and reduced response to non- familiar conspeciﬁcs (Supplementary Fig. 6). Thus, Nlgn3 expression, in both developing and mature VTA DA circuits, is required for habituation and nonfamiliar conspeciﬁc exploration.

A synaptic signature of saliency detection in VTA DA neurons.

Several experiences strengthen synaptic transmission at excitatory inputs onto DA neurons and drive the insertion of GluA2-lacking

Fig. 2VTA DA neuron excitability controls preference for nonfamiliar conspeciﬁc.aLeft: experimental time-course. Right: apparatus schematic and occupancy plot.bTime in chamber for vehicle treated VTA::DA^hM4Di. RM one-way ANOVA (chamber main effect:F(1.969, 33.48)=21.79,P< 0.0001,ﬁrst 5 mins; chamber main effect:F(1.881, 31.98)=2.825,P=0.0771, last 5 mins) followed by Holm-Sidak post hoc test.cTime in chamber for CNO treated VTA::

DA^mCherry. RM one-way ANOVA (chamber main effect:F(1.645, 23.03)=8.959,P=0.0022,first 5 mins; chamber main effect:F(1.545, 21.63)=3.665,P= 0.0527, last 5 mins) followed by Holm-Sidak post hoc test.dTime in chamber for CNO treated VTA::DA^hM4Dimice. RM one-way ANOVA (chamber main effect:F(1.494, 25.4)=5.248,P=0.0191,first 5 mins; chamber main effect:F(1.663, 28.27)=5.006,P=0.0182, last 5 mins) followed by Holm-Sidak post hoc test.eTime sniffing for vehicle treated VTA::DA^hM4Di. RM two-way ANOVA (stimulus main effect:F(1,34)=7.634,P=0.0092; time main effect:F(1,34)= 9.617,P=0.0039; time × stimulus interaction:F(1,34)=18.41,P=0.0001) followed by Bonferroni post hoc test.fTime sniffing for CNO treated VTA::

DA^mCherry. RM two-way ANOVA (stimulus main effect:F(1,28)=13.96,P=0.0008; time main effect:F(1,28)=5.028,P=0.0330; time × stimulus interaction:F(1,28)=5.629,P=0.0248) followed by Bonferroni post hoc test.gTime snifﬁng for CNO treated VTA::DA^hM4Dimice. RM two-way ANOVA (stimulus main effect:F(1,34)=1.458,P=0.2356; Time main effect:F(1,34)=4.806,P=0.0353; time × stimulus interaction:F(1,34)=0.6434,P=0.4281) followed by Bonferroni post hoc test.hDistance moved. One-way ANOVA (group main effect:F(2, 48)=12.86,P< 0.0001,ﬁrst 5 mins; group main effect:

F(2, 48)=15.01,P< 0.0001, last 5 mins) followed by Bonferroni post hoc test for planned comparisons.iSocial novelty index. RM two-way ANOVA (time main effect:F(2,48)=10.54,P=0.0021; group main effect:F(2,48)=35.503,P=0.0212; time × group interaction:F(2,48)=6.23,P=0.0039) followed by Bonferroni post hoc test for planned comparisons.Nindicates number of mice. Error bars represent s.e.m

(7)

AMPARs, which can be assessed by calculating a rectiﬁcation index (RI)

¹⁷

. We tested whether nonfamiliar exploration induced speciﬁc forms of long-lasting synaptic plasticity at excitatory inputs onto DA neurons in the VTA (novelty-induced synaptic plasticity). In WT mice, the RI increased at synapses 24 h after the exploration of either a nonfamiliar mouse or a nonfamiliar object

when compared to RI calculated from home caged mice (Fig. 6a).

By contrast, the RI was unchanged after the exposure to a new context and AMPA/NMDA ratios were unchanged for any of the above conditions (Fig. 6b). When AMPAR EPSCs were recorded after repeated exposure (over 4 days) to object stimuli, the RI was normalized to control condition (Fig. 6c). A subsequent exposure

e o1

e s1

f1 e

PosttestPretest

Familiar mouse

f1 ^s1Nonfamiliar mouse ^o1

d e

c

Pretest Posttest 0.0

0.2 0.4 0.6 0.8 1.0

Preference score

P = 0.0499

PreferenceAvoidance

N=10

0.2 0.4 0.6 0.8 1.0

Preference score

P = 0.0013

PreferenceAvoidance

N=10

0 50 100 150 200

Sessions

g

0 5 10 15 20 25

Sessions

h f

P < 0.0001 P = 0.0008

Cum. interaction (s)

i

Familiar mouse

f1 s1 ^o1 Nonfamiliar object

s1 o1

PreferenceAvoidance

0.2 0.4 0.6 0.8 1.0

Preference score

N= 10 P = 0.8394

0

1 2 3 4 5 6 7 8 9

10 11 12 1 2 3 4 5 6 7 8 9

10 11 12 10

20 30 40 50

Day 1 Day 4

Familiar mouse f1

0.2 0.4 0.6 0.8 1.0

Preference score PreferenceAvoidance

Pretest Posttest N=12 P = 0.0057 P = 0.6412 WT

group-housed Pretest

1

3

4 2

Posttest

Conditioning (day)

a b

PosttestPretest PosttestPretest

Sessions

DAT-Cre AAV5-DIO-hM4Di or

AAV5-DIO-mCherry Pre-TEST

1

3

4 2

Posttest

Conditioning (day)

j

PretestPosttest

VTA::DA^mCherry CNO

e s1

PretestPosttest

VTA::DA^hM4Di CNO

e s1

o1 e

o1 e o1 e f1 e s1 e

f1 e s1 e

f1 e s1 e f1 e s1 e 5′ 5′

x3

Nonf amiliar object Familiar mouse Nonf

amiliar mouse

4–6 wks

P = 0.0121 P < 0.0001 P = 0.0037 P < 0.0001

N=14

CNO CNO

s1 e CNO i. p.

Nonf amiliar mouse

Nonf amiliar mouse 5′ 5′

x3

0 100 200 300 400

Day 4 Day 1

f1 s1 o1 f1 s1 o1

N= 10 N= 10 N= 10

Day 1 Day 4 Day 1 Day 4

P = 0.0150

k

_{P = 0.0105}

VTA::DA^mCherry VTA::DA^hM4Di

Nonfamiliar mouse

Nonfamiliar object

Nonfamiliar mouse Nonfamiliar object

(8)

to a new object (o2) increased the RI (Supplementary Fig. 7a). By contrast, GluA2-lacking AMPARs were detected in mice repeat- edly exposed to a nonfamiliar conspecific stimulus (s1) over a 4- day period and were still present at these synapses after 10 days of repeated exposure (Fig. 6c). Remarkably, the AMPA/NMDA ratio was significantly elevated after 4 days of social (s1) repeated exposure relative to baseline but was normalized after 10 days of repeated exposure (Fig. 6d), while the paired-pulse ratio (PPR) remained unchanged throughout (Supplementary Fig. 7b). Taken together, these data indicate that repeated exposure to a non- familiar conspecific stimulus, but not an object stimulus, tran- siently increases synaptic strength (AMPA/NMDA ratio) and produces a stable insertion of GluA2-lacking AMPARs at VTA DA neuron excitatory inputs.

To understand the functional role of noncanonical AMPARs inserted during nonfamiliar conspecific exposure, we infused the GluA2-lacking AMPAR blocker NASPM into the VTA starting from the second day of interaction with either social or object stimuli (Fig. 7a, b). NASPM infused mice reduced the interaction with a conspecific stimulus upon repeated exposure (Fig. 7c); by contrast, the infusions did not alter long-term habituation to an object (Fig. 7d), interaction in the home cage between two familiar mice or distance moved in an open field (Supplementary Fig. 7c–e). To further understand the impact of GluA2-lacking AMPARs at VTA DA neuron inputs on conspecific repeated exposure, we promoted the insertion of GluA2-lacking AMPARs via blue-light illumination of ChR2 or eYFP expressing VTA DA neurons

¹⁸

of DAT-Cre mice (VTA::DA

^ChR2

: AAV5-Ef1α-DIO- ChR2(H134R)-eYFP, VTA::DA

^eYFP

: AAV5-Ef1α-DIO-eYFP).

DA neuron stimulation consisted in 15-minute long ChR2- mediated bursts of action potentials

¹⁸

delivered the day before each conspeciﬁc exposure (Fig. 7e, f). This noncontingent burst activation increased RI in photocurrent positive neurons (I

ChR2+

; Fig. 7g) and blocked habituation to social stimuli (Fig. 7h).

Altogether, these data indicate that GluA2-lacking AMPARs might represent a synaptic signature of conspeciﬁc saliency and, once inserted, their activity counteracts habituation.

Nlgn3 loss-of-function impairs novelty-induced plasticity.

Nlgn3 has been implicated in the regulation of AMPARs at glu- tamatergic synapses

³⁹

. We therefore hypothesized that defects in DA neuron synaptic function could represent the mechanism underlying the aberrant habituation to familiar conspeciﬁcs and response to nonfamiliar conspeciﬁcs in VTA::DA

^NL3KD

mice. We explored glutamate receptor function in VTA DA neurons of global Nlgn3

^KO

and conditional VTA::DA

^NL3KD

mice. Notably,

we observed increased RI of AMPAR-mediated currents indi- cating the aberrant presence of GluA2-lacking AMPARs at excitatory inputs onto VTA DA neurons in both Nlgn3 loss-of- function models (Fig. 8a). Given the abnormal elevation of GluA2-lacking AMPARs in naïve VTA::DA

^NL3KD

mice, we hypothesized that in these mice synaptic plasticity induced by exposure to nonfamiliar conspeciﬁcs might be occluded. Indeed, GluA2-lacking AMPARs in VTA DA neurons were not further increased 24 h after nonfamiliar social stimulus exposure in VTA::DA

^NL3KD

mice (Fig. 8b). Thus, aberrant plasticity of GluA2-lacking AMPARs in VTA DA neurons is associated with an impaired response to a social novel stimulus.

Discussion

In this study, we establish that intact VTA DA neuron excitability is necessary for (1) the exploration of nonfamiliar social stimuli, (2) the preference for nonfamiliar versus familiar conspecifics, and (3) the acquisition of nonfamiliar conspecific-induced con- textual associations. Novel stimuli, independent of their nature, leave a plasticity trace at glutamatergic synapses in the VTA, which persists upon repeated exposure to social stimuli and supports sustained conspecific interactions. We use a deletion of the ASD-associated gene Nlgn3 and demonstrate that global Nlgn3 knock-out results in an impaired habituation and an aberrant exploration of nonfamiliar conspecifics. Furthermore, selective inactivation of Nlgn3 in VTA DA neurons disrupts novelty-induced plasticity at glutamatergic synapses in the VTA, alters exploration of nonfamiliar conspecific, and the reinforcing properties of conspecific interactions while having no detectable effect on motor behaviors or olfaction.

Global loss of Nlgn3 is also accompanied by a broad spectrum of additional phenotypes, including changes in olfaction and in motor-related behaviors

^38–41

. Thus, the origin of social behavior alterations in these mice was unclear. Previous studies explored phenotypes in mice carrying a point mutation in Nlgn3 that reduces (but does not abolish) Nlgn3 expression and has been observed in 2 patients from one family

⁴³

. For this model, it was concluded that behavioral phenotypes are signiﬁcantly dependent on the genetic context with signiﬁcant phenotypes reported for some genetic backgrounds but not others

^44–46

. Our study demonstrates that VTA DA neuron specific Nlgn3 loss of func- tion is sufficient to recapitulate sociability deficits reported in global KO mice.

Although several studies have provided instrumental infor- mation about the neuronal circuits, within the reward system, that control social behavior in rodents

^13,47–49

, the synaptic

Fig. 3VTA DA neuron excitability mediates the reinforcing properties of nonfamiliar conspeciﬁc.aExperimental protocol for conditioned place preference with different stimuli.bRepresentative occupancy plots.cScatter plot of preference score measured for familiar mouse pairing during CPP. Pairedttest (t(9)=0.2086; mean and s.e.m for Pre-TEST: 0.498 ± 0.0298; mean and s.e.m for Post-TEST: 0.506 ± 0.0562).dScatter plot of preference score for nonfamiliar conspeciﬁc pairing during CPP. Pairedttest (t(9)=4.578; mean and s.e.m for Pre-TEST: 0.497 ± 0.0144; mean and s.e.m for Post-TEST: 0.596

± 0.0285).eScatter plot of preference score for novel object pairing during CPP. Pairedttest (t(9)=2.263; mean and s.e.m for Pre-TEST: 0.510 ± 0.0430;

mean and s.e.m for Post-TEST: 0.403 ± 0.0455).fTime course of interaction during conditioning blocks with a familiar mouse (f1). Friedman test (P= 0.0150;x²(12)=23.51).gTime course of interaction during conditioning blocks with a nonfamiliar conspeciﬁc (s1). Friedman test (P< 0.0001;x²(12)= 52.71).hTime course of interaction during conditioning blocks with a novel object (o1). Friedman test (P=0.0008;x²(12)=31.88).iCumulative interaction during conditioning sessions at day 1 and day 4, respe–Wallis test (K(6)=46.09,P< 0.0001) followed by Dunn’s test for planned comparisons.jLeft:

experimental protocol for VTA::DA^hM4Diand VTA::DA^mCherrytreated with CNO during CPP with nonfamiliar conspecific pairings. Right: representative occupancy plots for CNO treated VTA::DA^mCherryand VTA::DA^hM4Di.kScatter plot of preference score for VTA::DA^mCherrytreated with CNO during conditioning sessions with a nonfamiliar conspecific (mean and s.e.m for Pre-TEST: 0.4808 ± 0.0267; mean and s.e.m for Post-TEST: 0.5836 ± 0.0275), and scatter plot of preference score for VTA::DA^hM4Ditreated with CNO during conditioning sessions with a nonfamiliar conspecific (mean and s.e.m for Pre-TEST: 0.4903 ± 0.0162; mean and s.e.m for Post-TEST: 0.5091 ± 0.0428). RM two-way ANOVA (time main effect:F(1, 24)=7.7048,P=0.0105; virus main effect:F(1,24)=0.8678,P=0.3609; time × virus interaction:F(1,24)=3.2861,P=0.0824) followed by Bonferroni post hoc test for planned comparisons.Nindicates number of mice. Error bars represent s.e.m

(9)

adaptations occurring at VTA DA neurons during interactions with nonfamiliar conspecifics remained largely unknown. Here, we show that while reduced excitability or conditional suppres- sion of Nlgn3 in VTA DA neurons affect the exploration of nonfamiliar conspecifics, they both fail to modify responses to novel objects, presumably because of the higher intrinsic salience of social stimuli. Consistent with this hypothesis, we observe that while both nonfamiliar object and conspecific exploration trigger the insertion of GluA2-lacking AMPARs, only the repeated exposure to the same nonfamiliar mouse results in the

maintenance of GluA2-lacking AMPARs at glutamatergic synapses of VTA DA neurons. Therefore, we hypothesize that while the insertion of non-canonical AMPARs reﬂects the novelty associated to the stimulus, their persistence signals the higher salience of the conspeciﬁc over the object stimulus. The insertion and the expression of noncanonical AMPARs has been previously associated with nonsocial and highly salient experiences, such as cocaine exposure

¹⁷

. However, a causal relationship between behavioral responses to salient stimuli and GluA2-lacking AMPAR expression at VTA DA neurons has not been

Novel object recognition

f g h i j

a b

s1 s1 s1 s1 s2

c d e

5 min 5 min

1 h

1 2 3 4 5

Habituation Nonfamiliar Days 7–13 weeks

s1 s1 s1 s1 s2

15′ 15′ 15′ 15′ 15′

o1 o1 o1 o1 o2

or or or or or Time interaction (s)

Wild type (N= 12) Nlgn3^KO (N= 10) 0

50 100 150

10 P = 0.0221

0 50 100 150

–50 0

50 100 150

Time interaction (s)

Wild type (N= 12) Nlgn3 ^KO(N= 10) o1 o1 o1 o1 o2

Social novelty index (s2-s1#4)

0 50 100 150

–50

P = 0.0479

P = 0.0484

P < 0.0001 P < 0.0001

P = 0.0037

P = 0.0175

Nlgn3^KO

Wild type 0 10 20

30 ^{P = 0.0009} ^{P = 0.0014} 0.8 0.6

0.4

0.2

0

P = 0.6035

Velocity (cm/sec)

0 5 10

15 ^{P = 0.0047} Open field

0 5 10 15 20

Marble burying P < 0.0001

# marbles buried

N= 12 N= 12 10

8

N= 8 8 N= 10 12 N= 13 8

1 2 3 4 5

Time (day)

Wild type Nlgn3 KO

Time (day) 5

1 2 3 4

Object novelty index (o2–o1#4)

Wild type Nlgn3 KO

Discrimination ratio

N= 8

Wild type Nlgn3 KO

Wild type Nlgn 3^KO

Wild type Nlgn3 KO

k l

PreferenceAvoidance N = 17 N = 15

30′/30′

Cage-mates Empty

s e

Pretest

s e s e s e s e

1 2 5 6

Conditioning (days)

Posttest

30′/30′ 30′/30′ 30′/30′ Social CPP

P = 0.0440

Nlgn3^KO Wild type

Preference score

P = 0.0188 P > 0.9999

0.0 0.2 0.4 0.6 0.8 1.0

Pretest Posttest Pretest Posttest

Fig. 4Global knockdown ofNlgn3alters sociability and social reward behaviors.aExperimental time-course for the habituation/nonfamiliar exploration task.bMean social interaction time for wild type (WT) andNlgn3^KOmice. RM two-way ANOVA (time main effect:F(4, 80)=20.3,P< 0.0001; genotype main effect:F(1, 20)=3.629,P=0.0713; time × genotype interaction:F(4, 80)=6.071,P=0.0003) followed by Bonferroni’s post hoc test.cSocial novelty index of WT andNlgn3^KOmice. Unpairedttest (t(20)=2.481).dMean object interaction of WT andNlgn3^KOmice. RM two-way ANOVA (time main effect:F(4, 80)=17.07,P< 0.0001; genotype main effect:F(1, 20)=3.858,P=0.0636; time × genotype interaction:F(4, 80)=1.715,P=0.1547) followed by Bonferroni’s post hoc test.eDot plot of object novelty index for WT and Nlgn3^KOmice. Mann–WhitneyU=30.fSchematic of novel object recognition test.gTime spent investigating a novel and a familiar object. Pairedttest (WT:t(7)=5.494. Mean and s.e.m familiar=6.841 ± 1.41, mean and s.e.m novel

=14.33 ± 2.617. KO:t(7)=5.12. Mean and s.e.m familiar=8.115 ± 1.186, mean and s.e.m novel=16.63 ± 2.441).hObject discrimination ratio for WT and Nlgn3^KOmice. Unpairedttest (t(14)=0.5314).iMean velocity of WT andNlgn3^KOmice during a 7 min-openﬁeld test. Unpairedttest (t(20)=3.178).j Number of marbles buried for WT andNlgn3^KOmice. Unpairedttest (t(19)=5.505). (k) Schematics of the social conditioned place preference (CPP) test.l Scatter plot of preference score measured during the Pre- and Post-TEST for WT (mean and s.e.m for Pre-TEST: 0.4846 ± 0.0209; mean and s.e.m for Post-TEST: 0.5578 ± 0.0158), andNlgn3^KOmice (mean and s.e.m for Pre-TEST: 0.4809 ± 0.0178; mean and s.e.m for Post-TEST: 0.4886 ± 0.0225). RM two-way ANOVA (time main effect:F(1, 30)=4.422,P=0.0440; genotype main effect:F(1,30)=3.492,P=0.0715; time × genotype interaction:F(1,30)= 2.885,P=0.0998) followed by Bonferroni post hoc test for planned comparisons.Nnumbers indicate mice. All error bars are s.e.m

(10)

d

a b

AAV2-DIO-miRNlgn3-GFP or AAV2-GFP

P5-P6

P35-45

Social CPP P50

Open field/novel object recognition P60

Marble burying P80

Social nonfamiliar exploration VTA

SNc TH DIO-miRNlgn3-GFP HOECHST

8 weeks postinfection

g f

VTA::DA^NL3KD (N = 8) VTA::GFP (N = 14)

e

k

h i

c

j l

30′/30′

Cage-mates Empty

s e

Pretest

s e s e s e s e

1 2 5 6 Posttest

30′/30′ 30′/30′ 30′/30′ Social CPP Conditioning (days)

PreferenceAvoidance

0.8 0.6 0.4 0.2 0 1.0

Preference score VTA::DA^NL3KDVTA::GFP

N = 14 N = 8

1 2 3 4 5

Habituation NonFamiliar Days 7–13 weeks

s1 s1 s1 s1 s2

P = 0.0527

P = 0.0134 P > 0.9999

s1 s1 s1 s1 s2

0 50 100 150

P < 0.0001 P = 0.0237

P = 0.0002

0 50 100 150

–50

P = 0.0087

Novel object recognition

5 min 5 min

1 h

P = 0.0024 P = 0.0003

0 10 20 30 40

VTA::DA^NL3KD VTA::GFP

Discrimination ratio

0.8 0.6 0.4 0.2 0

1.0 P = 0.8724 _P = 0.5699

0 5 10 15 20 25

Velocity (cm/s)

P = 0.6742

0 5 10 15 20 25

Marble burying

# marbles buried

Open field 8 N = 14

8

N = 14 N = 14 8 N = 14 8 N = 14 8

Pretest Posttest Pretest Posttest

5

1 2 3 4

Time (day)

Social novelty index (s2–s1#4)

VTA::GFP VTA::DA

NL3KD

Familiar

Nonfamiliar VTA::GFP

VTA::DA NL3KD

VTA::GFP VTA::DA

NL3KD

VTA::GFP VTA::DA

NL3KD

Fig. 5Nlgn3in VTA DA neurons is required for social exploration and the reinforcing properties of conspeciﬁc interaction.aLeft: representative image of coronal slice of VTA and SNc from an AAV2 DIO-miR^Nlgn3-GFP infected DAT-Cre mouse. Right: higher magniﬁcation of VTA. Scale bar: 1 mm and 100μm.

bExperimental schematic of behavioral test order in VTA-injected mice.cExperimental schematic of the social-CPP test.dScatter plot of preference score measured during the Pre- and Post-TEST for VTA::GFP (mean and s.e.m for pre-TEST=0.4642 ± 0.0247. Mean and s.e.m post-TEST=0.5526 ± 0.0200), and VTA::DA^NL3KDmice (mean and s.e.m for Pre-TEST: 0.4434 ± 0.0218; mean and s.e.m for Post-TEST: 0.4548 ± 0.0214). RM two-way ANOVA (time main effect:F(1, 20)=4.24,P=0.0527; virus main effect:F(1,20)=6.103,P=0.0226; time × virus interaction:F(1,20)=2.527,P=0.1276) followed by Bonferroni post hoc test for planned comparisons.eExperimental schematic of the habituation/nonfamiliar exploration task.fMean social interaction plotted for VTA::GFP and VTA::DA^NL3KDmice. RM two-way ANOVA (time main effect:F(4, 80)=8.058,P< 0.0001, virus main effect:F(1, 20)=9.164,P= 0.0067; time × virus interaction:F(4, 80)=3.179,P=0.0178) followed by Bonferroni’s post hoc test.gSocial novelty index for VTA::GFP and VTA::

DA^NL3KDmice. Unpairedttest (t(20)=2.908).hExperimental schematic of novel object recognition test.iTime spent investigating a novel and a familiar object. Pairedttest (VTA::GFP:t(13)=3.763. Mean and s.e.m familiar=6.199 ± 0.805, mean and s.e.m novel=14.03 ± 2.188. VTA::DA^NL3KD:t(7)=6.518.

Mean familiar=8.226, s.e.m ± 1.069, mean novel=16.2, s.e.m ± 1.582).jDiscrimination ratio for object discrimination plotted for VTA::GFP and VTA::

DA^NL3KD. Unpairedttest (t(20)=0.1627).kMean velocity of VTA::GFP and VTA::DA^NL3KDmice during a 7 min openﬁeld test. Mann–WhitneyU=47.l Number of marbles buried plotted for VTA::GFP and VTA::DA^NL3KD. Mann–WhitneyU=49.5.Nnumbers indicate mice. All error bars are s.e.m

(11)

reported. Here, we show that the insertion of noncanonical AMPARs at VTA DA neurons contributes to behavioral responses to social stimuli suggesting that these receptors could also represent a functionally-relevant synaptic signature respon- sible for the behavioral responses associated with other salient stimuli.

Changes in AMPA/NMDA ratio occur in response to both rewarding and aversive processes

⁵⁰

, and synaptic strengthening is transiently expressed and necessary for associative learning

¹⁵

. Consistent with previous ﬁndings

¹⁶

, we report an increased AMPA/NMDA ratio at VTA DA neuron excitatory inputs in response to social interaction, which is transiently expressed upon repeated exposure to nonfamiliar conspeciﬁcs, but not object stimuli. However, although the increased synaptic strength might represent an additional signature related to the saliency of social interaction, its role in habituation processing and, possibly con- textual learning, warrants further investigation.

In recent years, accumulating evidences indicate that synaptic adaptations associated to reward and aversion occur at projection-speciﬁc subclasses of VTA DA neurons

^51,52

. An anatomic-functional segregation of reward circuitry is also emerging in respect of social behavior: while VTA DA neurons projecting to the nucleus accumbens (NAc), but not prefrontal cortex (PFC), promote conspeciﬁc interaction

¹³

, DA neuron projections to the interpenduncular nucleus control familiarity signaling

⁴⁷

. Thus, the specific synaptic signatures observed in response to nonfamiliar conspecific exposure might also occur in dedicated VTA circuits. At the same time, given the intrinsic diversity of sensory and emotional information provided by social vs. inanimate stimuli, it is conceivable that synaptic plasticity occurs at specific inputs to defined subclasses of VTA DA neu- rons. Additional investigations of synaptic properties of defined inputs to projection-specific DA neuron subclasses is needed to further understand the circuits and the synaptic mechanisms

0.0 0.5 1.0 1.5 2.0

Rectification index

P = 0.9510 P = 0.0352 P = 0.0162

P > 0.9999

a

1 (day) s1oro1 WT

Ncor

15′ 15′ 15′ 15′ 15′

15′ 15′ 15′ 15′ 15′ 15′ 15′ 15′ 15′ 15′

15′

B Ex vivo slice

physiology 24 h

BBaseline

(homecage) NcNonfamiliar

context s1 Nonfamiliar

mouse o1 Nonfamiliar object

B

Nc s1

o1

–60 mV +40 mV

B Nc o1 s1 8,3 12,3 11,4 12,4

b

1 (day) s1oro1 WT

Ncor

B Ex vivo slice

physiology 24 h

B Baseline

(homecage) NcNonfamiliar

context s1 Nonfamiliar

mouse o1 Nonfamiliar object

B

Nc s1

0.0 0.5 1.0 1.5 2.0

AMPA/NMDA

B Nco1 s1

o1

15,6 7,3 13,4 11,4 P = 0.9933

0.0 0.5 1.0 1.5 2.0

Rectification index

1 s1

o1 or

2 s1

o1 or

3 s1

o1 or

4 s1

o1 or Habituation WT

Ex vivo slice physiology 24 h

s1=s1x4 o1 = o1 x4 B

10 (day) s1

s1

–60 mV +40 mV

s1

–60 mV +40 mV B

o1

B o1s1 s1

s1=s1x10

P > 0.9999 P = 0.0167

P > 0.9999

c

0.0 0.5 1.0 1.5 2.0

AMPA/NMDA

B o1s1 s1 B

o1

s1 1

s1

o1 or

2 s1

o1 or

3 s1

o1 or

4 s1

o1 or Habituation WT

24 h

s1=s1x4 o1 = o1 x4 B

10 (day) s1

s1=s1x10

d

P > 0.9999 P = 0.0190 P = 0.0241

s1

Ex vivo slice physiology

18,4 10,3 28,7 12,3 13,4 8,3 20,6 9,3

(n,N)

Fig. 6Novelty-induced synaptic plasticity.aTop: experimental paradigm. Bottom: scatter plot of rectiﬁcation index and AMPAR-EPSCs example traces (−60, 0, and 40 mV) recorded from VTA DA neurons at baseline (B, homecage), or 24 h after 15 min of novel context (Nc), nonfamiliar conspeciﬁc (s1) or novel object (o1) exposure. One-way ANOVA (F(3, 39)=4.153,P=0.0120) followed by Bonferroni post hoc test for planned comparisons.bTop:

experimental paradigm. Bottom: scatter plot and example traces of AMPA/NMDA ratio recorded from VTA DA neurons at baseline (B, homecage), or 24 h after 15 min of novel context (Nc), nonfamiliar conspeciﬁc (s1) or novel object (o1) exposure. One-way ANOVA (F(3, 42)=0.0287,P=0.9933).cTop:

experimental paradigm. Bottom: scatter plot and example traces of rectiﬁcation index recorded from VTA DA neurons at baseline (B), 24 h after four repeated exposures to either a novel mouse (s1) or a novel object (o1) and ten repeated exposures to a nonfamiliar conspeciﬁc (s1, bold purple). One-way ANOVA (F(3, 64)=5.149,P=0.0030) followed by Bonferroni post hoc test for planned multiple comparisons.dTop: experimental paradigm. Bottom:

scatter plot and example traces of AMPA/NMDA ratio recorded from VTA DA neurons at baseline (b), 24 h after four repeated exposures to either a nonfamiliar conspeciﬁc (s1) or a novel object (o1) and ten repeated exposures to a nonfamiliar conspeciﬁc (s1, bold purple). One-way ANOVA (F(3,46)= 4.4939,P=0.0076) followed by Bonferroni post hoc test for planned multiple comparisons.n,Nindicates number of cells and mice respectively. Scale bars: 20 msec, 20 pA. Error bars report s.e.m